Analysis of Benchmark Forecasting Models Versus a Recurrent Neural Network Forecast of Channel Cargo Demand

Abstract

Given the longstanding history of inaccurate cargo demand forecast and underutilization of contracted commercial airlift augmentation, the primary objective of this paper is to present a simple forecasting model developed using Python. This study compares benchmark time series forecasting models against a Recurrent Neural Network model to determine which model is best for predicting monthly channel cargo airlift demand. To determine which model produces the best forecast, univariate time series analysis is conducted on Integrated Development Environment/Global Transportation Network Convergence data using readily available statistical modules and Machine Learning algorithms within the Python ecosystem.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 03, 2022
Accession Number
AD1177711

Entities

People

  • Abraham N Umanah

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Air Platforms
  • Autonomy
  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Air Force
  • Artificial Intelligence
  • Business Administration
  • Combatant Commanders
  • Computational Science
  • Congress
  • Department Of Defense
  • Financial Management
  • Governments
  • Information Science
  • Information Systems
  • Law
  • Machine Learning
  • Management Personnel
  • Mathematics
  • Military Operations
  • Military Personnel
  • Military Science
  • National Security
  • Neural Networks
  • Organizational Structure
  • Recurrent Neural Networks
  • Supervised Machine Learning
  • Technical Information Centers
  • Transportation
  • United States Central Command
  • United States Transportation Command

Readers

  • Aerospace logistics and air mobility.
  • Computational Modeling and Simulation
  • Distributed Systems and Data Platform Development

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks